Broadcast transmission system with sampling and correction arrangement for correcting distortion caused by amplifying and signal conditioning components

RELATED APPLICATION

This application is a continuation of U.S. patent application Ser. No. 09/105,824, filed Jun. 26, 1998, now U.S. Pat. No. 6,285,412, and U.S. patent application Ser. No. 09/312,344, filed May 14, 1999, now U.S. Pat. No. 6,281,936.

TECHNICAL FIELD

The present invention relates to broadcast transmission systems and is particularly directed to compensation of distortion within a digital transmission system, such as a digital TV (“DTV”) transmission system.

BACKGROUND OF THE INVENTION

A broadcast transmission system, such as a DTV broadcast system, includes an amplifying device that increases the power of an electrical information signal such that an antenna is excited to emit a broadcast signal at a desired strength. The amplifying device is referred to as a power amplifier. In order to optimize the quality of the broadcast signal, the electrical signal is conditioned prior to amplification. The signal conditioning includes band-pass filtering the electrical signal to limit the frequency band of the electrical signal that is input to the power amplifier.

Several issues arise during operation of such a transmission system. One issue is that the components of the transmission system, including the power amplifier and the signal conditioning devices, distort the electrical information signal away from intended values. Specifically, the power amplifier imposes non-linear distortion upon the signal. Also, some of the signal conditioning devices (e.g., band-limiting filter) impose linear distortions upon the information signal.

As a result of such distortions within the transmission system, instantaneous amplitude variations (AM/AM) and instantaneous phase variations (AM/PM) occur. In addition, frequency dependent amplitude and phase variations also occur. It is to be appreciated that within a phase-amplitude modulated system, amplitude and phase integrity of the system must be preserved for optimum system performance.

Traditional equalization for television systems has been accomplished by analog, pre-distortion equalizers and correctors that are static (non-adaptive). Such equalizers and correctors require factory adjustments to provide a desired amount of pre-distortion (pre-equalization). Aging of components, and temperature change cause drift in the proper amount of pre-distortion that should be imposed by the equalizers and correctors. Occasional field adjustments are required.

Digital signal processing techniques provide improved performance of the pre-distortion of the information signal. Specifically, digital signal processing can be used in an adaptive correction/equalization approach. Such an adaptive approach can eliminate the factory and field adjustments.

It is known to perform adaptive correction of a signal within a signal stream proceeding toward an antenna. However, in a relatively fast data system, the correction requires a relatively large amount of processing in a short period of time. In one known technique, all of the distortion (i.e., linear and non-linear) is corrected in a single step.

In another technique, the correction for the distortion imposed within the system is done component by component proceeding in a direction toward the antenna. Specifically, for each component, the signal that is output from that component is monitored to determine the amount of distortion imposed by that component. A correction is then developed for that component. Subsequently, the next component along the signal stream is monitored to develop the correction for that component. However, such a technique is time consuming and is often unsuitable for a high data rate stream. Thus, there is a need for a high-speed technique for adaptive correction of linear and non-linear distortion within a digital broadcast transmission system.

A second issue that presents itself is that the power amplifier may impose a frequency spectrum spread on the signal during amplification. The spreading may include smearing of the frequency and generation of unwanted frequency components. The frequency spread results in a broadcast signal of diminished quality. Additional signal conditioning, primarily in the form of band-pass filtering, after amplification will improve the quality of the broadcast signal. However, each additional signal-conditioning component (e.g., a band-pass filter) causes additional distortions to the signal. An increase in the number of distortion-causing components within the system is associated with an increase in the distortions that must be corrected. Practical signal processing systems have finite “real-time” processing capabilities and are subject to cost and complexity constraints. A system with a large number of distortion-causing components in a high data-rate system such as HDTV can easily exceed the capabilities of an implementable correction system.

SUMMARY OF THE INVENTION

The present invention provides a transmission system for broadcasting an information signal. The system includes signal provision means for providing an information signal. Power amplifier means amplifies the information signal to a broadcast transmission power level. The power amplifier means subjects the information signal to distortion shifts away from intended values. Pre-amp conditioning means conditions the information signal before the information signal is amplified by the power amplifier means. The pre-amp conditioning means is located upstream of the power amplifier means. The pre-amp conditioning means subjects the information signal to distortion shifts away from intended values. Post-amp conditioning means conditions the information signal after the information signal is amplified by the power amplifier means. The post-amp conditioning means is located downstream of the power amplifier means. The post-amp conditioning means subjects the information signal to distortion shifts away from intended values. First compensating means modifies the information signal to compensate for the distortion shifts imposed by the post-amp conditioning means. The first compensating means is located upstream of the pre-amp conditioning means. Second compensating means modifies the information signal to compensate for the distortion shifts imposed by the power amplifier means. The second compensating means is located downstream of the first compensating means and upstream of the pre-amp conditioning means. Third compensating means modifies the information signal to compensate for the distortion shifts imposed by the pre-amp conditioning means. The third compensating means is located downstream of the second compensating means and upstream of the pre-amp conditioning means.

BRIEF DESCRIPTION OF THE DRAWINGS

The foregoing and other features of the present invention will become apparent to one skilled in the art to which the present invention relates upon reading the following description of the invention with reference to the accompanying drawings, wherein:

FIG. 1 is a block diagram of an apparatus in accordance with the present invention;

FIG. 2 is a block diagram of an example device in which the present invention is utilized;

FIG. 3 is a flow chart of a process performed within the apparatus of FIG. 1; and

FIG. 4 is a flow chart of a correction/adaptation process performed within the apparatus of FIG. <b>1</b>.

DESCRIPTION OF A PREFERRED EMBODIMENT

One representation of the present invention is an apparatus <b>10</b> shown in function block format in FIG. 1 as a plurality of components that define a path of a data stream <b>12</b>. An information data signal proceeds along the data stream <b>12</b>. Preferably, the information signal has a relatively high data rate. The high data rate is related to the system environment in which the apparatus <b>10</b> is located. Specifically, the apparatus <b>10</b> is preferably part of a high definition (“HD”) digital television (“DTV”) system <b>14</b> as shown in FIG. <b>2</b>. Preferably, the DTV system <b>14</b> broadcasts signals in the radio range of frequencies. In one embodiment, the broadcast signal is in the ultrahigh frequency range (300-3000 MHz), and is preferably in the range of 470-860 MHz.

In pertinent part, the DTV system <b>14</b> includes an 8VSB exciter <b>16</b> and a transmitter <b>18</b>. The components of the apparatus <b>10</b> shown in FIG. 1 are located within the 8VSB exciter <b>16</b> and the transmitter <b>18</b> of FIG. <b>2</b>. Specifically, the transmitter <b>18</b> includes a power amplifier <b>20</b> (FIG. 1) that amplifies the information signal to a power level that is suitable for broadcast transmission of a RF signal. In one example, the amplified power level is 50 kilowatts. The power amplifier <b>20</b> may be comprised of an array of amplifying devices. If a plurality of amplifying devices is present within the power amplifier <b>20</b>, a combiner device is present to combine amplifier device outputs. It is to be understood that various amplifier configurations could be employed.

Turning now to the components located upstream of the transmitter <b>18</b> (FIG. <b>2</b>), many of these upstream components operate in digital format and at certain predetermined data rates. In particular, the 8VSB exciter <b>16</b> processes information digitally. Further, at one point within the 8VSB exciter <b>16</b>, a baseband modulator outputs the information signal in a complex domain, digital format, with an output sample rate equal to the baseband symbol rate. For HDTV, the rate of 10.76 Mega-samples per second (Msa/s).

In distinction, the power amplifier <b>20</b> amplifies an analog signal at a desired frequency to convey a relatively high rate of data. Thus, a series of components is located upstream of the power amplifier <b>20</b> to convert and condition the information signal to provide the desired input to the power amplifier. Specifically, (starting at the lower right corner of FIG. 1) a digital signal form of the information signal is provided at a predetermined data rate (e.g., 43.04 Msa/s) to a digital-to-analog converter (DAC) <b>24</b>.

The DAC <b>24</b> converts the information signal to analog form. The output frequency may be at any convenient intermediate frequency (IF). In the illustrated embodiment, the output signal frequency is centered at a frequency of 10.76 MHz. A low-pass filter <b>26</b> is located downstream of the DAC <b>24</b>. The output of the low-pass filter <b>26</b> is provided to a first up-converter <b>28</b> that is driven by a first local oscillator <b>30</b>. A band-pass filter <b>32</b> is interposed between the first up-converter <b>28</b> and a second up-converter <b>34</b>. A second local oscillator <b>36</b> drives the second up-converter <b>34</b>. The output of the second up-converter <b>34</b> is at the desired frequency and data rate for amplification by the power amplifier <b>20</b>.

A post-amplification filter <b>38</b> is located downstream of the power amplifier <b>20</b>. Herein, the post-amplification filter <b>38</b> is referred to as a high power filter <b>38</b>. The high power filter <b>38</b> is a band-limiting filter. It is to be appreciated that the transmitter <b>18</b> may include other components.

Focusing now upon a theoretical “ideal” system, all of the components of a transmitter of such an ideal system would be ideal. Specifically, a power amplifier of the system would be ideal and the transfer curve for the ideal amplifier would be linear. Thus, within such an ideal system, an information signal having a given pre-amplification power level would be amplified to a predetermined power level by the amplifier, based solely upon a linear relationship that dictates the amount of amplification. Also, filters of the ideal system would not impose any frequency dependent distortions.

The actual power amplifier <b>20</b> of the apparatus <b>10</b> is, however, not ideal. The actual power transfer curve of the power amplifier <b>20</b> is not linear. A non-linear distortion is imposed by the power amplifier <b>20</b> upon the information signal during amplification of the information signal. Specifically, the non-linear distortion is directed to changes in instantaneous amplitude and phase variations. Accordingly, a correction is desired upon the information signal to compensate for the distortion caused by the power amplifier <b>20</b>.

In addition, the filters of the transmitter <b>18</b>, and specifically the filters <b>26</b>, <b>32</b>, and <b>38</b>, impose linear frequency dependent deformations to the information signal. The low-pass filter <b>26</b> imposes a first linear distortion, the band-pass filter <b>32</b> imposes a second linear distortion and the high power filter <b>38</b> imposes a third linear distortion to the information signal. For example, the distortion imposed by the high power filter <b>38</b> is directed to group delay and amplitude response (i.e., amplitude variation versus frequency). Thus, for each distortion that occurs within the transmitter <b>18</b>, an amount of correction or equalization must be imposed upon the information signal to compensate.

Turning again to the theoretical ideal system, any action (i.e., amplification or filtering) imposed upon the information signal would be time-invariant. Specifically, in the ideal system, the actions imposed upon the information signal would not change over time. Thus, for a given input stimulus, the ideal system always produces the same output, independent of the time at which the stimulus occurs.

However, in actuality, the transmitter <b>18</b> is time-variant. Specifically, for a given input stimulus, the outputs of the components of the transmitter <b>18</b> change over time. One reason for time-variance is thermal effects within the transmitter <b>18</b>. The thermal effects cause variations in the amount of signal deformation caused by the power amplifier <b>20</b> and the filters <b>32</b> and <b>38</b> to the information signal. Thus, it is desirable to compensate for all of the signal distortion (i.e., the sequence of linear, non-linear, and linear), and adapt to changes in the distortion.

The apparatus <b>10</b> in accordance with the present invention provides three corrector or equalizer (i.e., compensating) components <b>42</b>-<b>46</b> within the 8VSB exciter <b>16</b> for the distortions that occur within the transmitter <b>18</b>. The corrector/equalizer components <b>42</b>-<b>46</b> are located upstream of the distorting transmitter components. Specifically, all of the corrector/equalizer components <b>42</b>-<b>46</b> are upstream of the DAC <b>24</b>. Thus, the correction/equalization is via pre-distortion of the information signal such that once distortion subsequently occurs at the transmitter <b>18</b>, the signal has desired values.

Turning to the specifics of the corrector/equalizer components <b>42</b>-<b>46</b>, an adaptive linear equalizer <b>42</b> imposes a pre-distortion onto the information signal to compensate for the linear distortion caused by the high power filter <b>38</b>. Preferably, the linear equalizer <b>42</b> includes at least one Finite Impulse Response (“FIR”) digital filter that has suitable structure for pre-compensating or pre-equalizing the information signal to compensate for the linear distortion caused by the high power filter <b>38</b>. The linear equalizer <b>42</b> may be comprised of, or include, a microprocessor that performs a program process and/or may be comprised of, or include, discrete “hard-wired” circuitry. It is to be appreciated that other filter types can be employed (e.g., IIR, a combination of FIR and IIR, or even an analog filter).

An adaptive non-linear corrector <b>44</b> imposes a pre-distortion onto the signal to compensate for the non-linear distortion caused by the power amplifier <b>20</b>. The non-linear corrector <b>44</b> may have any suitable structure for pre-distorting (i.e., pre-correcting) the signal to compensate for the non-linearities caused by the power amplifier <b>20</b>. Specifically, the non-linear corrector <b>44</b> may impose a linear piecewise correction curve and an iterative or empirical approach to routinely update a set of correction values within a memory. Alternatively, the correction could be generated by any number of algorithmic processes, such as curve fitting, that tend to provide the inverse distortion inherent in the power amplifier <b>20</b>. Thus, the non-linear corrector <b>44</b> may be comprised of, or include, a microprocessor that performs a program process and/or may be comprised of, or include, discrete “hard-wired,” or programmable circuitry.

An adaptive linear equalizer <b>46</b> imposes a pre-distortion onto the information signal to compensate for the pre-amplification linear distortion that is primarily caused by low-pass filter <b>26</b> and the band-pass filter <b>32</b>. Preferably, the linear equalizer <b>46</b> is a filter that has suitable structure for pre-compensating or pre-equalizing the information signal to compensate for the pre-amplification distortion. The linear equalizer <b>46</b> may be comprised of, or include, a microprocessor that performs a program process and/or may be comprised of, or include, discrete “hard-wired” or programmable circuitry.

The linear equalizer <b>42</b>, the non-linear corrector <b>44</b>, and the linear equalizer <b>46</b> are arranged in a sequence such that the pre-distortions (or pre-corrections) are imposed in a sequential order that is the inverse of the order that distortion occurs. Specifically, because the linear distortion caused by the high power filter <b>38</b> occurs last (i.e., at a downstream location from all of the other distortions), the pre-distortion imposed by the linear equalizer <b>42</b> occurs first. The pre-distortion imposed by the non-linear corrector <b>44</b> occurs second because the non-linear distortion imposed by the power amplifier <b>20</b> occurs second. The pre-distortion imposed by the linear equalizer <b>46</b> occurs third (i.e., after the pre-distortion from the linear equalizer <b>42</b> and the pre-distortion of the non-linear corrector <b>44</b>) because the pre-amplification linear distortion occurs prior to the distortion caused by the power amplifier <b>20</b> and the high power filter <b>38</b>.

The linear distortion caused by the high power filter <b>38</b> must be corrected first (i.e., prior to non-linear correction) such that frequency dependent variations do not impact the non-linear pre-distortion. Such a sequence avoids a problem that a correction is deficient, or even incorrect and in a direction opposite to the direction needed for proper correction. Accordingly, in the correction scheme in accordance with the present invention, the linear effects (such as group delay) of the high power filter <b>38</b> are corrected first. Thus, the amplitude and group delay variations over frequency are not misinterpreted as non-linear deformations to the information signal.

Turning to the signal input provided for the pertinent portion of the apparatus <b>10</b> shown in FIG. 1, the information signal that is output from the baseband modulator (i.e., complex, digital, and preferably at 10.76 Msa/s) is input to an interpolation component <b>54</b>. The interpolation component <b>54</b> interpolates the complex data stream by two (2) to 21.52 Msa/s. A converter <b>48</b> converts the information signal from complex format to real format, and also effectively doubles the sample rate of the information signal (preferably to a rate of 43.04 Msa/s). The output of the complex-to-real converter <b>48</b> is the input to the linear equalizer <b>42</b>. Thus, it is to be appreciated that the corrector/equalizer components <b>42</b>-<b>46</b> are located such that all of the correction/equalization occurs at baseband or at a relatively low IF.

As a digression regarding operation of the corrector/equalizer components <b>42</b>-<b>46</b> at this data rate, it is to be noted that the non-linear characteristics of the power amplified <b>20</b> generic spectral spreading. Prior to the power amplifier <b>20</b>, the signal bandwidth is confined ideally to a bandwidth set by a Nyquist filter (shaping filter) that is located upstream of the components shown in FIG. <b>1</b>. The non-linear distortion generated in the power amplifier <b>20</b> is broad-band, and extends beyond the Nyquist signal bandwidth. As the non-linearity becomes higher order, the bandwidth gets larger. As examples, a 3^rdorder non-linearity generates spectral spreading of about three times the original bandwidth, and a 5^thorder system equates to about five times the bandwidth.

Since the non-linear corrector <b>44</b> is required to correct for all of this spectral splatter, it must be capable of generating correction over the same bandwidth. This means that the signal applied to the non-linear corrector <b>44</b> must be over sampled by the same amount as the order of the non-linearities that are to be corrected. The components <b>54</b> and <b>48</b> allow three times the bandwidth of correction, and thus permit correction of 3^rdorder artifacts.

Turning again to the structure of the apparatus <b>10</b>, it is to be appreciated that the linear equalizer <b>42</b> operates on the signal in the real domain. A real-to-complex converter <b>50</b> is located between the linear equalizer <b>42</b> and the non-linear corrector <b>44</b>. Thus, the non-linear corrector <b>44</b> operates in the complex domain so that both amplitudes and phase correction can be accomplished. A complex-to-real converter <b>52</b> is located between the non-linear corrector <b>44</b> and the linear equalizer <b>46</b>. The linear equalizer <b>46</b> and the components of the transmitter <b>18</b> operate in the real domain. It is to be appreciated that FIG. 1 shows one embodiment of the linear equalizers. Because the linear equalizers <b>42</b> and <b>46</b> are real filters, there must be complex-to-real conversions <b>48</b> and <b>52</b> to allow for real operations on the signal. A real-to-complex converter <b>50</b> is needed to return the signal to a complex format preparatory to non-linear correction. If complex equalizers are used, then no complex-to-real pre-conversions and real-to-complex post-conversions are needed.

As stated above, the amount of correction/equalization imposed by the linear equalizer <b>42</b>, the non-linear corrector <b>44</b>, and the linear equalizer <b>46</b> can be adapted (i.e., updated). A controller <b>60</b> determines the amount of change of the correction/equalization for each of the linear equalizer <b>42</b>, the non-linear corrector <b>44</b>, and the linear equalizer <b>46</b> (e.g., the filter coefficients are changed). In order to make determinations regarding correction/equalization adaptation, the information signal is sampled prior to each correction/equalization component. The signal sample taken prior to the linear equalizer <b>42</b> is held within a W memory <b>62</b>. The signal sample taken prior to the non-linear corrector <b>44</b> is held within a D memory <b>64</b>. The signal sample taken prior to the linear equalizer <b>46</b> is held within an X memory <b>66</b>. In turn, the memories <b>62</b>-<b>66</b> are connected to the controller <b>60</b> to provide the signal sample values to the controller <b>60</b>.

Determinations of whether a correction/equalization requires adaptation (i.e., change) require comparisons between the information signal prior to the correction/equalization and the information signal after distortion occurs. Thus, samples of the information signal are taken for each distortion. Specifically, the information signal is coupled-off <b>70</b> just prior to the power amplifier <b>20</b>, such that the linear distortion of the band-pass filter <b>32</b>, etc. is discernable. The information signal is coupled-off <b>72</b> just after the power amplifier <b>20</b>, such that the non-linear distortion of the power amplifier <b>20</b> is discernable. The information signal is coupled-off <b>74</b> just after the high power filter <b>38</b>, such that the linear distortion of the high power filter is discernable.

A sampler <b>76</b> selectively samples at one of the three available sample locations (i.e., pre-amp, post-amp, and post-high power filter). The sampler <b>76</b> includes a switcher and a down converter. The output of the sampler is passed, via a low-pass filter <b>78</b>, to an analog-to-digital (A/D) converter <b>80</b> and then to an Y memory <b>84</b>. The Y memory <b>84</b> is connected to the controller <b>60</b>.

The controller <b>60</b> controls the sampler <b>76</b> to sample one of the three available sample locations (i.e., pre-amp, post-amp, and post-high power filter). The determination of which of the sample locations if chosen is dependent upon the correction/equalization that is to be monitored/adapted. The Y memory <b>84</b> thus holds the information signal values that are indicative of the distortion that is needed to make the adaptation determinations. Thus, less processor capacity is required because the controller <b>60</b> selectively chooses the distortion to monitor and correct at each moment, and the processing that does occur is at a reduced rate.

A process <b>100</b> for controlling the sampler <b>76</b> is shown in FIG. <b>3</b>. The process <b>100</b> begins at step <b>102</b> and proceeds to step <b>104</b>, in which the sampler <b>76</b> awaits a switch instruction from the controller <b>60</b>. At step <b>106</b> the controller <b>60</b> provides a switch instruction. At step <b>108</b>, the sampler <b>76</b> adjusts its switch setting according to the instruction from the controller <b>60</b>. The information signal is sampled (step <b>110</b>) at the chosen “pick-off” location (i.e., pre-amp, post-amp, or post-high power filter). The process <b>100</b> goes to step <b>112</b> to determine if the controller <b>60</b> requests a change (i.e., change or disable the sampler). If the determination at step <b>112</b> is negative (i.e., the controller has not provided a new command) the process loops to step <b>110</b> and the signal continues to be sampled at the chosen pick-off location. If the determination at step <b>112</b> is affirmative (i.e., the controller has provided a new command) the process goes to step <b>104</b> to perform the steps <b>104</b>-<b>108</b> for switch adjustment.

A process <b>200</b> for correction/adaptation is shown in FIG. <b>4</b>. The process <b>200</b> begins at step <b>202</b> and proceeds to step <b>204</b>, in which the linear equalizer <b>42</b> is set to provide a predetermined amount of compensation. Preferably, the initial compensation provided by the linear equalizer <b>42</b> is a nominal high-power filter compensation. At step <b>206</b>, the non-linear corrector is initialized to provide an initial predetermined correction. Preferably, the initial correction provided by the non-linear corrector is a nominal power amplifier correction. At step <b>208</b>, the linear equalizer <b>46</b> is initialized to provide a predetermined compensation. Preferably, the initial compensation is a nominal sinx/x and up-converter compensation.

At step <b>210</b>, the sampler is set to sample at <b>70</b>. At step <b>212</b>, the X and Y memories are filled. Linear equalization is provided at equalizer <b>42</b> based upon a comparison of the values in the X and Y memories at step <b>214</b>. At step <b>216</b>, the sampler is set to <b>72</b>. The D and Y memories are filled at <b>218</b>. The correction of the non-linear corrector <b>44</b> is optimized based upon a comparison of the values in the D and Y memories at step <b>220</b>. The sampler is set to <b>74</b> at step <b>222</b>. The linear equalizer <b>42</b> is equalized based upon a comparison of the values in the W and Y memories at step <b>224</b>. Upon the completion of step <b>224</b>, the process <b>200</b> loops back to step <b>210</b>.

From the above description of the invention, those skilled in the art will perceive improvements, changes and modifications. Such improvements, changes and modifications within the skill of the art are intended to be covered by the appended claims.